This randomized, placebo-controlled, double-blind, multicenter clinical trial will assess the efficacy of adjunctive cyclic progesterone therapy in lessening the frequency of intractable seizures in women with localization-related epilepsy. There is considerable scientific evidence to suggest that estrogen generally increases while progesterone decreases neuronal excitability and seizures. Preliminary open trials suggest that the cyclic administration of adjunctive natural progesterone supplement may lessen seizure frequency by over 50% in the majority of women with catamenially exacerbated intractable seizures. Oral synthetic progestins, in contrast, have not shown significant efficacy. Progesterone is not widely prescribed because its benefits have yet to be conclusively demonstrated. We have shown the feasibility and safety of a progesterone trial, using 150 subjects in the initial phase of this investigation. We now propose proceding with the completion of this definitive investigation of 640 subjects. During the baseline phase, seizure and menstrual charts document baseline seizure frequency and determine if seizure occurrence shows a catamenial pattern of exacerbation. The subjects are then divided into catamenial and non-catamenial strata. Each stratum is randomized in a 2:1 ratio into progesterone and placebo treatment groups. Seizure frequency during 3 months of treatment is compared to baseline, while monitoring antiepileptic drug and hormone levels. This clinical trial has considerable potential significance for women with epilepsy. Approximately 30% have persistent seizures despite antiepileptic drug use. It is estimated that 35% of these women have catamenial seizure exacerbation. If this group responds favorably to hormonal therapy, one would expect that progesterone may benefit approximately (1,000,000 x .30 x .35) 100,000 women with catamenial epilepsy and perhaps many more women with no a priori demonstrated hormonal sensitivity to seizure occurrence.